| FazBrowse GitHub Viewer | Trending | | Home |
| Tools: [Download Repo ZIP] [Original HTTPS Page] |
Sorry, something went wrong.
|
Removed flaky status for the test. Rerun CI just because: https://ci.nodejs.org/job/node-test-pull-request/664/ |
Sorry, something went wrong.
|
Well, this is a bummer, but it looks like this does not fix the pipeflood test problem on Windows: https://ci.nodejs.org/job/node-test-binary-windows/189/RUN_SUBSET=1,VS_VERSION=vs2015,label=win2012r2/tapTestReport/test.tap-228/ |
Sorry, something went wrong.
|
Although it's failing differently so maybe it can be tweaked... |
Sorry, something went wrong.
|
Wrapping the req.socket.on('data',...) business in setImmediate() (so that any events already in the event queue will fire first?), the test works reliably on Windows now. With it wrapped and set to run 16 times, no failures:
With it not wrapped and set to run 16 times, failures:
With changes to get rid of ES6-isms, etc., Node v0.10.20 fails on the test still and node v0.10.21 passes, which is as it should be. |
Sorry, something went wrong.
Sorry, something went wrong.
|
Now it's failing on the Pi. (╯°□°)╯︵ ┻━┻ https://ci.nodejs.org/job/node-test-binary-arm/431/RUN_SUBSET=1,nodes=pi2-raspbian-wheezy/console |
Sorry, something went wrong.
|
Switched to common.platformTimeout() instead of hardcoding the same timeouts for arm as for everything else. CI run looks better. https://ci.nodejs.org/job/node-test-pull-request/686/ Only failures are for other flaky tests and build-bot failures. |
Sorry, something went wrong.
There was a problem hiding this comment.
I vaguely remember a PR to make all Errors be called with new. May be worthwhile to make it consistent?
Sorry, something went wrong.
There was a problem hiding this comment.
@evanlucas Sure, we can do that. .... Done!
Sorry, something went wrong.
There was a problem hiding this comment.
"the mechanism" or "the flood prevention mechanism"?
Sorry, something went wrong.
|
I think this test could live in test/parallel but I'd appreciate it if you did the move in a separate commit. |
Sorry, something went wrong.
|
OK, made all the changes suggested by @bnoordhuis, rebased, force pushed. PTAL |
Sorry, something went wrong.
|
CI: https://ci.nodejs.org/job/node-test-pull-request/708/ Hope it doesn't have any surprises... |
Sorry, something went wrong.
This extends fixes for test-https-pipeline-flood to hopefully fully eliminate its flakiness on Windows in our continuous integration process. PR-URL: #3636 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
|
landed in v4.x-staging in 5254fda...9294523 |
Sorry, something went wrong.
|
Would be nice to have this landed in 0.12.X as it is the last persistent failure on windows that keeps builds from being green |
Sorry, something went wrong.
PR-URL: #3636 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
This extends fixes for test-https-pipeline-flood to hopefully fully eliminate its flakiness on Windows in our continuous integration process. PR-URL: #3636 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
PR-URL: #3636 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
This extends fixes for test-https-pipeline-flood to hopefully fully eliminate its flakiness on Windows in our continuous integration process. PR-URL: #3636 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
PR-URL: #3636 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
This extends fixes for test-https-pipeline-flood to hopefully fully eliminate its flakiness on Windows in our continuous integration process. PR-URL: #3636 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
This extends fixes for test-https-pipeline-flood to hopefully fully eliminate its flakiness on Windows in our continuous integration process. PR-URL: nodejs#3636 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
|
@Trott I'm wondering why its set as dont-land-on-v4.x and the lts-watch-v0.12 was removed. Sounds like we are not backporting but not clear why from what I've read in the issue. |
Sorry, something went wrong.
Sorry, something went wrong.
| Back | FazBrowse Home | New Git URL |
Continuation of #2862